UNIVERSITY OF PISA
FACULTY OF MEDICINE AND SURGERY
SCHOOL OF PhD DEGREE IN
CLINICAL PATHOPHYSIOLOGY AND DRUG SCIENCES
PROGRAM IN
PATHOPHYSIOLOGY AND MEDICAL
President: Prof. Eleuterio Ferrannini
PHARMACOLOGY
PREVALENCE OF SECONDARY HYPERPARATHYROIDISM
IN DIALYSIS PATIENTS: COMPARISON BETWEEN
BIOCHEMICAL AND MORPHOLOGICAL
DIAGNOSIS
PhD thesis: Ilaria Petrucci, MD
Supervisor: Mario Meola, MD PhD.
INDEX
ABSTRACT...2
INTRODUCTION...4
ANATOMY AND EMBRYOGENESIS OF PARATHYROID GLANDS…...4
Anatomy…...4
Embryogenesis...6
PHYSIOPATHOLOGY OF SECONDARY HYPERPARATHYROIDISM...7
CLINICAL CONSEQUENCES OF SECONDARY HYPERPARATHYROIDISM….10 Uremic osteodystrophy...10
Extra-skeletal consequences...12
TREATMENTOFSECONDARYHYPERPARATHYROIDISM ... 16
Calcimimetics: a new era ... 19
DIAGNOSISOFSECONDARYHYPERPARATHYROIDISM Serum levels of i-PTH: biochemical diagnosis of sHPT...24
... 23
Imaging techniques ... .... 26
BACKGROUND OF THE STUDY ... 29
AIM OF THE STUDY ... 31
PATIENTS AND METHODS ... 32
Patients ... 32 Study design ... 32 Methods ... 32 Statistical analysis ... 35 RESULTS ... 36 DISCUSSION ... 42 CONCLUSIONS ... 50 REFERENCES ... 52
ABSTRACT
Patients with Chronic Kidney Disease (CKD) can develop secondary hyperparathyroidism (sHPT), which is an adaptive response triggered by persistent alterations of the mechanisms that control the homeostasis of calcium (Ca), phosphorus (P) and vitamin D. In epidemiological studies, the prevalence of sHPT is based on mineral metabolism biochemical parameters and on serum intact parathyroid hormone (i-PTH) levels. Based on these parameters, the prevalence varies from 20-30% to 47-50%.
The primary objective of this study was to establish the
epidemiology of a cohort of maintenance hemodialysis (MHD) and peritoneal dialysis (PD) patients. Secondary objectives of the study were: 1) to integrate the biochemical and pharmacological data with the morphological data obtained by Ultrasonography with colorDoppler (US/CD) and 2) to establish the correlation between morphological and biochemical sHPT. We enrolled 395 patients, 269 males and 126 females, 380 in MHD and 15 in PD. All patients underwent US/CD examination of the thyroid/parathyroid glands. The examination was performed by two experienced operators; single-blind (ICC 0.95-intrapersonal, interpersonal-ICC 0.91). Epidemiological data were collected in a case report format. Mean dialytic age of population studied was 62.9 ± 71.3 months while mean predialytic CKD age was 91.4 ± 82.7 months. Serum Ca was 8.9 ± 0.8 mg / dl, P 4.7 ± 1.5 mg / dl, Ca x P product 41.8 ± 15.0 mg2/dl2
The prevalence of biochemical sHPT was 36.2% (143 patients out of 395) while the prevalence of morphological sHPT was 27.3% (108 patients out of 395).
, ALP 133.5 ± 93.1 IU/l and i-PTH 281.9 ± 233.4 pg/ml.
According to the reference values recommended by the K-DOQI guidelines, 252 patients showed i-PTH serum levels ≤300 pg/ml (144.1 ± 78.9 pg/ml) and 143 patients had i-PTH serum levels >300 pg/ml (526.6
± 215.9 pg/ml). Serum Ca levels did not differ significantly while P, Ca x P product and ALP were significantly different (P 4.6 ± 1.4 mg/dl vs 4.9 ± 1.7 mg/dl, p = 0.03, Ca x P 40.5 ± 13.8 mg2/dl2
US/CD showed 173 parathyroid glands in 108 patients: 70 (64.9%) of them showed values of PTH >300 pg/ml and 38 (35.1%) values of i-PTH ≤300 pg/ml. 73/287 (25.4%) patients showed values of i-PTH >300 pg/ml, without hyperplastic parathyroid glands at US/CD.
vs 44.4 ± 16.6 p = 0.01, ALP 124.7 ± 87.2 IU/l vs 148.9 ± 101.1 IU/l, p=0.01).
The correlation between i-PTH and gland volume showed a linear trend (R2 = 0.1365, p = 0.001) regardless of current treatment (glandular volume = 0.8995 * (i-PTH) + 15.3). The biochemical evaluation associated with morphological examination with US/CD is the most complete diagnostic protocol for sHPT. So if i-PTH is constantly >400-500 pg/ml, an US/CD evaluation of the neck should be performed in order to document the presence of hyperplastic glands.
INTRODUCTION
Secondary hyperparathyroidism (sHPT) develops progressively during the course of chronic kidney disease (CKD). Phosphorus retention and reduced synthesis of 1–25 vitamin D3
Ultrasound with colorDoppler (US/CD) is the first choice imaging technique for the location and volume assessment of parathyroid glands, for the diagnosis of diffuse or nodular hyperplasia, as well as for clinical assessment of sHPT.
result in low serum calcium levels, which causes a feedback secretion of intact parathyroid hormone (iPTH). Long-term hyperstimulation of parathyroid glands enhances cell proliferation resulting first in diffuse polyclonal hyperplasia and finally in monoclonal nodular hyperplasia.
ANATOMY AND EMBRYOGENESIS OF PARATHYROID GLANDS Anatomy
Parathyroid glands are small endocrine glands with variable size and shape. They are located, two on each side (upper and lower parathyroid glands), on the posterior medial edge of the thyroid lobes. The average size of a normal gland is 5 x 3 x 1 mm, while the average weight is about 35 mg [1]. Glandular shape is often oval or spherical, but may even be oblong, lobed or multilobar. The parathyroid parenchyma of a young adult has a low amount of fat tissue and adipose stromal cells in normal glands are characterized by intracellular fat vacuoles at microscopic analysis.
Parathyroid glands have an arterial pedicle that penetrates into the gland at the hilum, where it can be divided into two or three branches. The parathyroid artery origins from the inferior thyroid artery in most cases (80 to 90%). The inferior parathyroid gland can be feed by a branch from the arch of the aorta (middle thyroid artery), from the brachiocephalic trunk or from the ipsilateral internal mammary artery. The venous drainage is ensured by a subcapsular network which flows towards the hilum. Superior
parathyroid glands veins drain to the middle thyroid veins, while the inferior glands drain mostly in the inferior thyroid veins [2].
The microscopic structure of parathyroid glands is characterized by strings of endocrine cells surrounded by connective tissue. Chief cells are the most represented cells. They are small (12-20 microns in diameter) and round, with a central nucleus and a cytoplasm that may look clear, pale or dark, depending on their functional status. Among chief cells, the most represented in normal conditions are the clear cells. The chief cells release parathyroid hormone (PTH). The chief cells biological activity is regulated by extracellular levels of calcium (Ca2+
Normal parathyroid cells are characterized by a very low turnover [6]. They show a very slow rate of mitosis (slow apoptosis). Their average life span was estimated by Parfitt et al. using the nuclear antigen Ki-67. In human adults is approximately 20 years. [7]. Several studies have shown that regeneration does not exceed 5% of the cell population/year [8,9]. A recent analysis of samples of human parathyroid glands with the TUNEL technique showed an apoptosis rate of 1/10000 cells [10,11].
). Finally, parathyroid parenchymal cells have oxyphil and transitional cells. These are slightly larger than chief cells, with a smaller central core and a cytoplasm rich in mitochondria and glycogen granules, but with very rare secretory vesicles, so they appear eosinophilic and easily detectable. The functional role of oxyphil cells has not yet been fully elucidated, although these cells have been recently identified as responsible for the synthesis of an i-PTH-related protein – (PTHrP) [3] that could act as an autocrine/paracrine regulator of proliferation [4,5]. In old subjects, fat tissue progressively replaces glandular cells, while in the newborn and infancy glands are composed almost entirely by chief cells.
Several autoptic studies [2,12-16] show that the anatomical distribution of the parathyroid glands is relatively constant, especially for the superior glands located in most cases above the intersection between the recurrent laryngeal nerve and the inferior thyroid artery. In fact, superior
parathyroid glands are located in 77% of cases near the cricothyroid junction, above the intersection between the recurrent laryngeal nerve and the inferior thyroid artery, they are surrounded by fat tissue and show a long vascular pedicle; in 22% of cases they are rear of the upper pole of the ipsilateral thyroid lobe. In 1% of cases they are located in the retropharyngeal or retroesophageal space. The inferior parathyroid glands have a wider distribution than the superior ones. These glands are situated between the lower pole of the thyroid gland and the thymus and, rarely, if they have not migrated during the embryonic development, they are higher in the neck, close to the jaw. Only in rare cases parathyroid glands were found in the pericardial sac. The inferior parathyroid glands are located in 42% of cases near the anterior or posterolateral surface of the inferior pole of the thyroid gland. Here, parathyroid glands are frequently masked by vessels or by the thyroid tissue. In 39% of cases, they are contained in the fat tissue between the lower pole of thyroid gland and the thymic horn or they are close to the thymus in the superior mediastinum. In 15% of cases, then, they are distributed in the iuxtathyroid space, lateral to the inferior thyroid pole. Other more rare ectopic locations are the mediastinum, the intrathyroid or prethyroid, retropharyngeal and retroesophageal space. In only 8% of cases the inferior glands can be above the junction between the laryngeal nerve and the inferior thyroid artery.
Embryogenesis
The knowledge of the embryonic development of the parathyroid glands is essential for the diagnosis and surgical management of hyperparathyroidism. It explains the different glandular locations, especially those related to an abnormal embryological migration which occurs in 2-5% of cases and leads to surgical failure. Parathyroid glands originate from the interaction of neural crest mesenchyme and third and fourth branchial pouch endoderm. The superior parathyroids arise from the fourth pharyngeal pouch, and the inferior parathyroids arise from the third pharyngeal pouch together with the thymus. Inferior parathyroid glands in the adult can be
located from the corner of the jaw to the area near the thymic space in the anterior superior mediastinum. Superior parathyroids maintain a more cranial and constant location, close to the posterior edge of the thyroid lobe, near the inferior thyroid artery [17].
PHYSIOPATHOLOGY OF sHPT
In literature, the prevalence of sHPT in haemodialysis patients is highly variable, from 20-30% [18] to 47-50% [19,20].
In normal conditions, parathyroid cells are quiescent with a low grade of mitotic divisions [8,21]. Therefore, many biochemical stimuli determine a rise of i-PTH secretion by an immediate release of preformed vesicles (in few minutes), then they determine the transcription of i-PTH-mRNA (few hours) and finally they induce a modification of the replication rhythm.
These stimuli are:
1. Hypocalcemia and a high ionized calcium set-point (ie the average
concentration of Ca2+
2. Reduced synthesis of 1-25 (OH)
that inhibits by 50% the secretion of PTH). In patients with sHPT the calcium-PTH sigmoidal curve is shifted to the right, so higher serum calcium levels are required to suppress the secretion of i-PTH [22,23]. Changes in the calcium-PTH balance and calcium set-point appear late in the course of CKD and they characterize patients with severe sHPT [24]. Furthermore, hypocalcemia is responsible for a post-transcriptional regulation of i-PTH [25].
2 vitamin D3 in the kidney [26]. Hyperphosphatemia, the loss of renal mass and an increase in FGF-23 reduce the activity of α1-hydroxylase. Consequently, decreased serum levels of 1-25 (OH)2 vitamin D3 determine a reduced intestinal absorption of calcium, which contributes to hypocalcemia and to an insufficient activation of the parathyroid glands receptors for the 1-25 (OH)2 vitamin D3 (Vitamin D Receptors - VDR), resulting in an increased transcription of PTH mRNA and then increase in its synthesis [27].
3. Hyperphosphatemia. According to the “trade-off” theory, the retention
of phosphate induces a reduction in serum Ca2+ levels.
4. The reduced density of calcium membrane receptors (calcium-sensing
receptors - CSR) [26] of parathyroid cells.
Experimental animal studies have also documented a direct action of phosphorus on parathyroid cells [28]: hyperphosphatemia is responsible for an increase in the PTH gene transcription, acting at a post transcriptional level [25]. Hyperphosphatemia also represents the main inhibitory stimulus for vitamin D activity.
5. The reduced concentration of receptors for the 1-25 (OH)2 vitamin
D3
6. The increase of Fibroblast Growth Factor 23 (FGF-23) serum levels.
FGF-23 is a 251 amino acids protein that in humans is encoded by the
FGF23 gene. FGF23 is a member of the fibroblast growth factor (FGF)
family which is responsible for phosphate metabolism. (calcitriol) [29].
Recent studies have shown a positive correlation between the progressive decline of renal function and the gradual increase in serum intact FGF-23 [30-32]. In hemodialysis patients FGF-23 is markedly increased, demonstrating a positive correlation with PTH and phosphorus. FGF-23 is a reliable screening test to identify patients who, within two years, will develop refractory secondary hyperparathyroidism [33]. Kazama JJ et al. have attribute to FGF-23 the additional role of marker for the resistance to treatment with intravenous calcitriol in patients with sHPT [34]. FGF-23 has the following functions: a) it inhibits the activity of tubular Na-P cotransporters, decreasing the reabsorption of phosphate and thus increasing its excretion; b) it reduces the expression of renal 1 α-hydroxylase inducing, therefore, a negative feedback on 1-25 (OH)2 vitamin D3; c) it induces the synthesis of 24-hydroxylase that degrades 1-25 (OH)2 vitamin D3 [35]; d) FGF-23 directly inhibits, in vivo and in vitro, the expression of PTH gene and the secretion of PTH [36,37]. The mechanism of action of FGF-23 is not yet completely clear, but it seems that it acts
through a specific receptor (FGF-R) expressed in the kidney, parathyroid and pituitary gland and choroid plexus. However, the affinity of FGF-23 to its receptor is quite low so a cofactor (Klotho protein) is needed. The effects of this protein are not fully elucidated, but it seems to be involved in the processes of aging.
In uremic patients, parathyroid cells are under continuous stimulation by hyperphosphatemia, hypocalcaemia, reduced synthesis of vitamin D3
The relationship between gland volume and indices of hormonal secretion is also well known [41]. In fact, the basal calcium-independent secretion of PTH positively correlates with the glandular size [42]. Serum i-PTH levels become able to overcome the skeletal resistance, thus determining a bone resorption. Despite the increase in serum calcium and phosphorus, parathyroid resistance to negative feedback of calcium allows i-PTH secretion to progressively increase. According to international guidelines, parathyroidectomy becomes mandatory when one or more parathyroid glands are enlarged in volume (>500 mm
and reduced expression of VDR and CSR. Long-term hyperstimulation of parathyroid glands produces glandular hyperplasia which is polyclonal at first and then monoclonal and nodular, progressing into a growth disorder (tertiary hyperparathyroidism) [38,18]. The evolution from diffuse polyclonal to nodular monoclonal hyperplasia/hypertrophy is accompanied by a further reduction in the density of CSR and especially of VDR; this different gene expression is responsible of the resistance to medical therapy with calcitriol [39]. Nevertheless, the pathogenesis and the progression of sHPT have a multifactorial etiology. Other etiologic and still unknown factors (probably genetic mutations) may play a role. In fact, several deletions of chromosome 11 have been identified in parathyroid glands of patients with tertiary hyperparathyroidism [40].
3
), i-PTH >700 pg/ml and the parameters of the mineral metabolism are no longer controlled by conventional therapy [43-45].
CLINICAL CONSEQUENCES OF sHPT
The lack of control on mineral metabolism and on serum i-PTH has important skeletal and extra-skeletal consequences in uremic patients. In the first group we include fibrocystic osteitis, mixed uremic osteodystrophy and adynamic bone disease, and the second group includes cardiovascular and soft tissue calcifications, endocrine abnormalities, impaired immune system, erythropoiesis abnormalities and neuromuscular disease. sHPT plays a very important role in the pathogenesis of cardiovascular disease which is the leading cause of morbidity and mortality in the uremic population.
Uremic osteodystrophy
The term “uremic osteodystrophy” includes all skeletal diseases in patients with CKD, characterized by an accurate histo-morphometric study of bone biopsy. Almost all uremic patients show heterogeneous histological bone abnormalities, divided into two groups: bone diseases with high or low turnover. Diseases with high bone turnover are fibrocystic osteitis and mixed uremic osteodystrophy. Low turnover bone diseases are osteomalacia and adynamic bone disease. The prevalence of the different forms of osteodystrophy has changed in the last decades: mixed uremic osteodystrophy and adynamic bone disease (ABD) are predominant nowadays, while osteomalacia has almost disappeared and has been replaced by ABD. Normally, the risk of a high bone turnover disease increases with the increase of i-PTH serum levels [46,47]. However, the ability to predict that there is a high bone turnover disease is poor until i-PTH serum levels do not reach 450-500 pg/ml. Instead, i-i-PTH serum levels <100 pg/ml are suggestive of a low bone turnover disease.
High turn-over bone disease. Fibrocystic osteitis is characterized by a high
rate of bone formation and resorption, an increased osteoclast and osteoblast activity and a progressive increase in peritrabecular marrow fibrosis. The bone erosion can occur in any district, but phalanges and skull are the most
common locations. Sever fibrocystic osteitis may be associated with brown tumors, which are pseudotumoral lytic lesions with haemorrhage and necrotic phenomena and it is associated with a high incidence of pathological bone fractures, especially vertebral.
Mixed uremic osteodystrophy is characterized by some aspects of a high turnover osteodystrophy associated with mineralization defects and an increase in bone formation. The hystological abnormalities show different degrees in every patient, but in all of them there is a coexistence of fibrotic areas with increased bone remodeling and deposition of low-mineralized bone matrix [48].
Low turn-over bone diseases. ABD is nowadays the most prevalent form
of bone disease associated with CKD in haemodialysis patients. The histologic features of ABD are the absence of osteoblastic and osteoclastic activity and a poor production of bone matrix [48]. An excessive suppression of i-PTH due to the use of phosphate binders and calcium-based dialysis baths containing high concentrations of calcium inhibit a bone structure which already has a very low turnover, as it occurs in haemodialysis patients. Therefore, skeletal structures are unable to absorb serum calcium and bones appear more fragile. The etiology remains unknown but risk factors are age, diabetes mellitus and excessive use of phosphate binders containing calcium and vitamin D. In addition, osteoblasts show an abnormal response to i-PTH due to downregulation of the PTH receptor [49], which further contributes to the poor cellularity that characterizes ABD.
Symptoms associated with renal osteodystrophy usually appear with
severe CKD. Bone and joint pain progressively worsen and they finally determine walking inability. Itching is due to the deposition of calcium in renal failure, especially in patients with severe hyperparathyroidism [50,51].
Foley [52] has demonstrated that the risk of cardiovascular mortality in dialysis patients is approximately 10-20 times higher than in the general population, particularly in younger subjects.
Vascular calcification. Vascular calcification is a physiopathological
mechanism that can explain the high mortality. Although all forms of uremic osteodystrophy are associated with vascular mineralization, the association between low-turnover osteodystrophy and arterial calcification is important: in fact, skeletal osteoblast function is deficient in ABD, but there is an osteoblastic differentiation of intimal-medial muscle cells that surround the atherosclerotic plaque. Vascular calcification is associated with four parameters in dialysis patients: hyperphosphatemia, hypercalcemia, elevated Ca x P product and oral intake of calcium.
1. Hyperphosphatemia, a major independent risk factor for cardiovascular disease, directly stimulates the calcification process and it induces the proliferation of smooth muscle cells [53]. Block et al. [54] demonstrate that over 60% of dialysis patients have a phosphatemia >5.5 mg/dl and patients with values >6.5 mg/dl have a higher risk of mortality by 27% compared with subjects with phosphorus ranging from 2.5 to 6.5 mg/dl.
2. There is a linear relationship between serum calcium levels and
cardiovascular mortality [55], probably linked to the calcium availability for the mineralization process [56]. The clinical consequences of these observations are clear if we consider that, except for calcimimetics, all therapeutic strategies that reduce PTH serum levels, such as vitamin D and calcium based phosphate binders, are often associated with an increase in serum calcium and Ca x P product [57].
In the general population, vascular disease may be due to different pathological processes associated with calcification: atherosclerotic disease and Mönckeberg medial calcific sclerosis. In atherosclerotic disease, characterized by wall fibro-fatty plaques, calcification is a late event [58]. However, atherosclerosis can also be a circumferential lesion (with no obstruction of the lumen) with calcification occurs earlier in the course of
the disease [59]. Mönckeberg calcific sclerosis of small and medium elastic arteries shows circumferential calcifications that appear like thin rings at X-ray scan [60]. Recent studies have clearly demonstrated that this form of medial calcification in distal vessels is associated with increased cardiovascular mortality in diabetic patients without chronic renal failure, [61] and in patients with chronic renal failure [62].
Recently it has been suggested that both vascular diseases have a unique physiopathological process [63]. In fact there is a dynamic process, not just a degenerative mechanism due to a passive deposition of calcium. The first step is the dedifferentiation of smooth muscle cells into osteoblast-like cells. These osteoblast-osteoblast-like cells produce type I collagen and non collagenous matrix proteins, which can regulate the mineralization process. The increase of CaxP product, sHPT or an excessive calcium intake may accelerate this process. Following to this theory, mineralization begins when promoting factors exceed inhibiting factors [64].
Calcified uremic arteriolopathy (CUA) of small vessels of skin, subcutaneous fat, and skeletal muscle determines ulcerative lesions that progress toward ischemic necrosis. In patients with CUA the mortality rate ranges from 45 to 65% and even higher in presence of ischemic lesions.
Further studies have shown structural and functional myocardial abnormalities, including fibrosis and calcification, associated with hyperphosphatemia and excessive calcium load in hemodialysis patients [65].
Role of PTH and cardiovascular risk in sHPT. The wide range of cellular
and tissue effects associated with increased serum PTH led to consider it as a uremic toxin [66]. High serum i-PTH levels (>65 pg/ml) in CKD patients are associated with a high risk of cardiovascular mortality, regardless phosphatemia and vitamin D therapy [67,68]. i-PTH has specific receptors on myocytes, and this direct effect causes an increase in intracellular calcium concentrations and rises the frequency and strength of muscle contraction [69]. Finally, it determines an increased interstitial collagen
tissue and a myocyte hypertrophy [70]. Left ventricular mass (LVM) and LVM indexed to height (LVMI) are higher in patients with hypertension and high serum levels of i-PTH [71]. PTH is also able to play an active role in the deposition of calcium in the myocardium, and in valvular and vascular structures [72-74].
Metastatic calcifications. The systemic consequences of sHPT must be
summed to all clinical consequences of the uremic state. Calcium-phosphorus metabolism and bone turnover abnormalities are associated with cutaneous, renal, pulmonary and cardiovascular metastatic calcifications, which are expression of the reduced bone formation, i.e. an excessive skeletal remodeling with increased serum levels of Ca, P and Ca x P product [75]. Heterotopic calcifications may occur in many soft tissues: cornea (band keratopathy), conjunctiva (red-eye syndrome), lung (restrictive syndrome), myocardium (arrhythmias, valvular calcifications), articular and periarticular structures, and tendons (joint pain and functional impairment, partial or complete rupture of the quadriceps muscle tendons or of the Achilles tendon).
Other clinical consequences of sHPT. sHPT is also involved in glucose
intolerance because i-PTH decreases the production of insulin, by increasing the concentration of intracellular calcium in pancreatic islet cells [76].
Some studies have also shown an immunosuppressive action of i-PTH
[77] with decreased neutrophil phagocytic capacity and susceptibility to infections. The increased concentration of intracellular calcium reduces the synthesis of ATP on the mitochondrial respiratory chain [78]. Moreover, there are also functional abnormalities of macrophages, immunoglobulins, platelets and T lymphocytes [76].
Patients with severe sHPT show a refractory anemia which requires higher doses of recombinant erythropoietin (EPO). This refractory anemia, not responsive to EPO, significantly improves after parathyroidectomy [78-80]. The main cause of this reduced treatment response is osteodystrophy
and bone marrow fibrosis with consequent reduction of erythroid cells [80]. However, studies have also suggested a possible direct toxic effect of i-PTH on erythropoietic bone marrow [81].
High i-PTH and intracerebral calcium serum levels are responsible for some manifestations of uremic encephalopathy, but mechanism involved are still unknown [82]. Some experimental data suggest that i-PTH interferes with normal neurotransmission, by promoting the entry of calcium in the central nervous system [83].
TREATMENT OF SECONDARY HYPERPARATHYROIDISM
Treatment of sHPT has two objectives: a. short-term objectives (maintenance of optimal values for i-PTH, calcium, phosphorus, prevention of parathyroid hyperplasia and of skeletal abnormalities); b. long-term objectives (reduction of cardiovascular morbidity and mortality).
In most cases, the reference ranges given by the K/DOQI guidelines for iPTH, P, Ca serum levels and CaxP product in haemodialysis patients are not achieved, as reported by the European multicenter observational analysis (COSMOS) based on 4500 patients [84] and the recent Italian Survey of the FARO Study Group based on 2637 patients [85]: only 30% of patients reaches the target for i-PTH and 50% for calcemia and phosphatemia.
Biochemical parameters Range
i-PTH 150-300 pg/ml
Ca 8,4-9,5 mg/dl P 3,5-5,5 mg/dl Ca X P <55 mg2/dl2
KDOQI guidelines reference ranges for i-PTH, P, Ca e CaXP product in haemodialysis patients. I-PTH: 1 pmol/l =1 pg/ml x 0,105; Ca: 1 mmol/l = 1 mg/dl x 0,25;P: 1 mmol/l = 1 mg/dl x 0,323.
Current treatment strategies rely on three drug families, based on the pathophysiological mechanisms of sHPT:
1. Phosphate binders. They are used in 90% of hemodialysis patients [86], because dialytic clearance and intake reduction of phosphorous (without causing malnutrition) are not sufficient alone to control hyperphosphatemia.
Aluminium salts (AlOH3
The K/DOQI guidelines [89] suggest that the maximum amount of calcium (used as a phosphate binder) must not exceed 1500 mg per day, while the K-DIGO new guidelines [90] do not set a specific limit, but they recommend to reduce the amount of calcium-based binders in case of hypercalcemia, vascular calcification, adynamic bone disease or low serum i-PTH levels. Sevelamer has been the first calcium and aluminium free phosphate binder that shows the same efficacy of calcium carbonate in controlling hyperphosphatemia, but at the same time it reduces the
) are probably the most effective phosphate binders in reducing the intestinal absorption of phosphate. However, other effects of aluminium salts as potential neurotoxicity of aluminum, anemia and its accumulation at the site of osteoid mineralization, which results in severe osteomalacia and ABD, have limited their use in clinical practice. Calcium salts (carbonate and acetate) are frequently used, and they are well tolerated but there is a risk of a positive calcium balance, hypercalcemia in near 50% of patients (especially when it is used at high doses and in association with vitamin D analogues), and vascular calcifications [87]. They can also determine an excessive suppression of i-PTH, increasing the risk of ABD [88].
Lanthanum carbonate is a salt formed by lanthanum cations and carbonate anions and it is used in Europe since 2006. It has demonstrated a high gastrointestinal phosphate binding capacity and a higher efficacy to that of sevelamer and of calcium carbonate. Its efficacy means fewer pills to take so it increases the patient compliance towards binding therapy.
2. Calcium supplements. The administration of calcium supplements
(calcium carbonate or calcium acetate) may be effective in controlling the i-PTH serum levels in some dialysis patients and in some cases of CKD stage
5. The use of calcium is limited due to the increased risk of vascular calcification.
3. Activators of the VDR (VDRA). Calcitriol therapy in sHPT inhibits the synthesis of parathyroid hormone and the parathyroid hyperplasia [91], and it is used in clinical practice since the '80s, but its use is accompanied by the risk of hypercalcemia and hyperphosphatemia determine due to the increase in intestinal calcium and phosphorus absorption, which determines the development or progression of vascular calcifications, especially if there is an excessive suppression of bone turnover [92].In the last decade, in order to minimize such negative effects, many new molecules have been introduced in drug therapy (paricalcitol, dorxecalciferol, alphacalcidol). They determine a selective activation of VDR and they maintain the inhibitory activity on the parathyroid gland but have minor effects on the intestinal transport of calcium and phosphorus, due to a different affinity for the vitamin D binding protein and the nuclear receptor. Paricalcitol is used in Europe for intravenous injection in hemodialysis patients, although it was also approved for oral administration even in the early stages of CKD. Some experimental studies suggest that paricalcitol is more effective than calcitriol in sHPT: a) minor influence on serum calcium and phosphorus [93]; b) different effects on cardiovascular tissues in terms of inhibition of proliferation of vascular smooth cells [94] and reduced induction of the processes of vascular calcification [95], suppression of the renin-angiotensin system, inhibition of myocytes proliferation; c) reduced risk of ABD [96]. It is however important to remember that there is a reduction of VDR in patients with parathyroid nodular hyperplasia, so vitamin D/analogues in patients with severe sHPT are less effective [97].
The definition of severe sHPT is based on the inadequate control of serum levels of Ca, P, Ca XP and i-PTH with conventional therapy associated with a significant increase in cardiovascular risk [98]. Following international guidelines, in fact, parathyroidectomy is mandatory when conventional therapy does not control mineral metabolism parameters, as well as the finding of a gland volume >500 mm3 and values of i-PTH >700
pg/ml [99,100]. Many observational studies highlight the failure to achieve and/or to maintain with conventional treatment the targets indicated by the K-DOQI guidelines [101,102] and they suggest the necessity of adding calcimimetics to the conventional therapy.
Calcimetics: a new era
Calcimimetics are a family of ligands that mimic or enhance on CaSR the effects of extracellular Ca2+
Pharmacodynamics. The biological activity of calcimietics is due to
the interaction with the transmembrane part of the receptor (6
. Class I includes calcimimetics that directly stimulate the receptor interacting with its extracellular domain. Calcimimetics of class II (NPS R-467, S-467, R-568, S-568, AMG 073) are allosteric activators of CaSR, which in parathyroid cells is the main regulator of the rapid secretion of preformed PTH [103].
th
and 7th
PTH decreases soon after administration of cinacalcet, reaching the nadir about 2-6 hours later, at the Cmax of cinacalcet. Later, while reducing the levels of cinacalcet, PTH levels increase up to 12 hours after administration, then the suppression of PTH remains approximately constant until the end of the dosing interval (24 hours). Once the steady state has been reached, serum calcium levels remain constant over the dosing interval.
transmembrane domain) that induces some conformational changes of the extracellular portion (N-terminal) of the receptor itself. The binding of the calcimimetic with CaSR increases the sensitivity of the receptor to extracellular calcium and therefore it determines a shift to the left of the sigmoidal curve, which describes the relationship ionized calcium-PTH. This action determines a reduction of the calcium set point and finally an inhibition of PTH secretion [104]. Cinacalcet (AMG 073) is the only second-generation calcimimetic used in clinical practice since 2004 in the USA and since 2005 in Europe and it is able to reduce i-PTH and calcium serum levels, without increasing the intestinal absorption of calcium and phosphorus. This characteristic distinguishes it from calcitriol and similar vitamin D analogues [105].
Pharmacokinetics. Based on comparisons between different studies,
it was estimated that cinacalcet has an absolute bioavailability in fasting subjects that reaches approximately 20-25%. The administration of cinacalcet with food determines an increase in the bioavailability of approximately 50-80%. Increases in plasma concentration of cinacalcet are similar, regardless of the fat content of foods. After absorption, cinacalcet concentrations decline in a biphasic way, with an initial half life of about 6 hours and a terminal half life of 30-40 hours. The steady-state drug levels are achieved within 7 days, with a very low accumulation. The AUC and Cmax of cinacalcet increase in a linear way in the dose range (from 30 to 180 mg once daily). The absorption is saturated at doses above 200 mg, probably due to a poor solubility. The pharmacokinetics of cinacalcet does not change over time. The volume of distribution is about 1000 liters, indicating a large distribution. Cinacalcet is bound to plasma proteins for about 97%. Cinacalcet is metabolized by multiple enzymes, primarily CYP3A4 and CYP1A2. The major circulating metabolites are inactive. Cinacalcet is rapidly and extensively metabolized by oxidation followed by conjugation. The renal excretion of metabolites is the predominant route of elimination of cinacalcet. There are no clinically relevant differences in the pharmacokinetics of cinacalcet due to age. The pharmacokinetic profile of cinacalcet in patients with CKD stage 1,2,3,4 and those on hemodialysis or peritoneal dialysis is similar to that found in healthy volunteers. Mild hepatic dysfunction does not affect the pharmacokinetics of cinacalcet. Compared to subjects with normal liver function, the mean AUC of cinacalcet is approximately 2 times higher in subjects with mild dysfunction and approximately 4 times higher in patients with severe hepatic dysfunction. In patients with mild and severe hepatic dysfunction the cinacalcet half life is longer respectively 33% and 70%. The protein binding of cinacalcet is not affected by liver disease. Since the dose is titrated for each subject based on the parameters of safety and efficacy in patients with liver dysfunction any further adjustment of the dose is not needed. The
clearance of cinacalcet is higher in smokers than in non smokers, probably due to induction of CYP1A2-mediated metabolism.
The cumulative analysis of 3 studies in phase 3 (1136 dialysis patients treated for 26 weeks) has demonstrated the effectiveness of the use of cinacalcet vs placebo in the control of sHPT, with a reduction in PTH of 40% vs 6%, which was associated with a reduction (6.8%) of serum calcium, of serum phosphorus (8.4%) and of CaXP product (15%) [106]. The OPTIMA randomized prospective study [107] showed that cinacalcet added to conventional therapy determines a significant increase in the percentage of patients that reach the target for i-PTH, calcium and phosphorus (according to the K-DOQI guidelines): 71% of patients in the group treated with conventional therapy and cinacalcet reaches a value of i-PTH <300 pg/mL compared to 22% of the control group. The stimulation of CaSR also inhibits the synthesis of i-PTH with a post-transcriptional mechanism [108], and inhibits the proliferation of glandular cells [109].
Calcimimetic also determines an up-regulation of CaSR and VDR on parathyroid cells [110,111], which potentially improves the therapeutic response to calcitriol. The EVOLVE clinical trial will assess the effects of the improvement obtained with cinacalcet on biochemical parameters of mineral metabolism and on the clinical outcomes, such as mortality and cardiovascular events in dialysis patients. The aim of another prospective randomized study in progress on 330 subjects (ADVANCE) is to test the hypothesis that treatment with cinacalcet attenuates the progression of vascular calcification in dialysis patients, as suggested by experimental studies in animals [112]. Moreover, cinacalcet seems to improve the histological bone turnover and to reduce the excessive marrow fibrosis in most patients with sHPT [113]. The hypothesis derived from preclinical studies [114,115] about the possibility that cinacalcet could induce regression of parathyroid hyperplasia has been also evaluated in humans [116]. Cinacalcet has a more rapid effect on biochemical profile of severe sHPT than on the morphology of the gland. In fact, the control of biochemical parameters is reached quickly, while the changes in volume,
ultrasonographic and vascular pattern appear only after 12-18 months. The most significant volume changes, however, are found in parathyroids with a starting volume <500 mm3, while in larger glands (>500 mm3
Cinacalcet modulates the biology of parathyroid cells in synergy with vitamin D and phosphate binders, and leads to the appearance of degenerative phenomena and glandular hypovascularisation. The pathophysiological mechanisms responsible for cystic degeneration and reduction of blood supply may find an explanation in an increased apoptosis [117,118], although it is not possible to exclude an indirect effect of cinacalcet on the vascular tree due to an interaction with endothelial CaSR [119]. Persistent receptor stimulation is also needed to achieve the control of mineral metabolism. In fact, the secretory rebound, appeared after the temporary suspension of cinacalcet, could be explained by the persistence of a functioning cell mass which is strongly stimulated by the suspension of the drug. The therapeutic response should be therefore monitored not only with biochemical but also with morphological parameters. It is also ) prevail involutive phenomena with the appearance of areas of cystic degeneration. In most cases, the ultrasonographic pattern changes from hypoechoic (suggestive of glandular hyperplasia with increased number of secreting cells) towards hyperechoic (cells become quiescent with an increase number of oxyphil cells and of adipocytes). It's interesting to observe that the glands showed a reduction in blood flow after treatment with cinacalcet regardless of the starting volume, and they reproduced in a reverse way the progression towards parathyroid hyperplasia (from hypoechoic to hyperechoic ultrasonographic pattern, and from hypervascularized to non vascularized color Doppler pattern). In any case, the most important morphological changes coincided with the most marked reduction of i-PTH and calcium serum levels, suggesting a major responsiveness in these cases. In some patients with these characteristics a lower dose than the minimum dose currently available (30 mg/day) was needed. It would be important in the future to start cinacalcet earlier and with lower doses, in combination with conventional therapy.
important to perform a long-term therapy to achieve a "pharmacological parathyroidectomy" that could reduce the number of patients who undergo surgical parathyroidectomy.
DIAGNOSIS OF SECONDARY HYPERPARATHYROIDISM
Diagnosis of biochemical sHPT is based on the evaluation of changes in calcium-phosphorus balance: hypocalcemia, hyperphosphatemia, decreased 1-25 (OH)2 vitamin D3 and increased i-PTH. In fact, there is a linear relationship between non-suppressible i-PTH secretion and glandular volume [41]. Imaging techniques allow understanding the clinical aspects of sHPT, thus reaching two purposes: a. anatomical location, number and morphology of hyperplastic parathyroid glands, and b. clinical condition of the target organ damaged by the imbalance of mineral metabolism (arteries, heart, bones, muscles, skin). However, the morphological diagnosis of sHPT may present several difficulties: 1. elevated i-PTH serum levels are not always synchronous with volumetric growth of parathyroid glands; 2. imaging techniques for the study of parathyroid glands (US/CD ultrasonography, scintigraphy, CT, MRI)may present difficulties of implementation and interpretation of the results.
Serum levels of i-PTH: biochemical diagnosis of sHPT
The dosage of parathyroid hormone remains an open question, due to the variability of several factors:
1. Biological variability. In healthy subjects and in patients with
primary hyperparathyroidism, PTH secretion is pulsatile, but in sHPT tonic and pulsatile secretion are increased and amplified as well as the frequency of concentration peaks. Other variable factors are age (significant increase with age in both sexes (+36%)), and familiarity [120].
2. Procedural variability. In order to maintain the sample stability, it
must be considered that serum is stable up to 6 hours at 20°C before being frozen, and if it must be preserved for longer periods it is recommended to use plasma EDTA (but i-PTH serum values result 10-20% higher).
3. Analytical variability. Immunoassays use antibodies against the
molecule to determine the PTH serum concentrations, so the identification of the antigen, and in particular the epitopes to which antibody is being developed, is crucial. For PTH, the problem is that there are several circulating molecules and they have different biological activities.
The whole molecule of PTH is constituted by 84 amino acids, but a rapid hydrolysis in the parathyroid gland produces both C-terminal and N-terminal fragments. Proteolysis can occur in different positions and it produces many different fragments. C-term fragments are considered to be biological inactive, while the N-term fragments (the most important is 1-34) are biologically active, and show a very short half-life. All these fragments are present in the serum sample for assay of PTH and, depending on the method used, they are detected or not. In the past it was believed that the biological effects of PTH resulted only from binding to specific receptors for PTH with affinity only for the whole molecule and the N-term fragment 1-34. But nowadays, it is known that many C-term fragments may inhibit the activity of PTH, by acting on receptors expressed on osteoblasts and osteocytes [121]. This has had important consequences in the routine dosage methods of PTH, leading to the proposal of new methods.
Assay methods. The first methods for the determination of PTH (first
generation essays), have been completely replaced by the immunoradiometric assay of "second generation" that uses the technique of the "double antibody". The capture antibody is directed toward the C-term portion of the molecule (35-84), while the tracer antibody is directed toward the soluble N-term portion (7-34). In recent years, the IRMA methods were replaced by immunochemiluminometric methods (ICMA), which do not use radioactive markers. Third generation methods detect 1-84 PTH or "whole" PTH. They have been introduced because second generation assays had an
important weakness: the tracer antibody was directed toward the 7-34 portion of the molecule but it was demonstrated that 30-50% of circulating PTH consists of N-truncated fragments of the molecule in positions 1-9, and in particular the fragment 7-84, biologically inactive and cross-reagent with the assays for intact PTH [122]. The values obtained with the whole PTH assay are 40-50% lower than the ones for intact PTH, both in CKD and healthy patients. This is ascribed to the interference due to the N-truncated fragments in position 1-7. At the moment, however, there is no recommendation to replace second-generation assays with those of third-generation [121].
Reference values and interpretative criteria. At present, the
reference range proposed by K-DOQI guidelines has been set with the kit “Allegro” intact PTH assay (II generation IRMA method), for which the most widely used commercial kits showed fluctuations in the value of the same sample between -43% and +123%. From a practical point of view, it is mandatory to standardize the method by considering the sampling method and the kit used. Clinicians have two ways to consider the PTH values obtained by their laboratories: the first one is to have specific ranges for each kit used, while the second is to use correction factors to compare it with the reference values of K-DOQI guidelines. In the first case, the UK Renal Association recommends to maintain i-PTH levels below four times the upper limit of the reference method used in patients with CKD, or after kidney transplantation, hemodialysis or peritoneal dialysis for at least three months. K-DOQI guidelines (2003) recommend that i-PTH must be ranged between 150 and 300 pg/ml.
Imaging Techniques
Ultrasonography (US) of the cervical region is the first imaging technique in the study of the parathyroid glands due to its repeatability, safety and low cost, although it needs to be performed a nephrologist which is expert in ultrasound. The data in the literature about sensitivity of US in sHPT are controversial, so they confirm the strong dependence between
results reliability and operator experience [123,124]. In a cohort of 40 patients undergoing parathyroidectomy for severe sHPT, Jeanguillame et al. reported a sensitivity of 54%. Ultrasonography with ColorDoppler (US/CD) allows the assessment of volumetric progression over time (changes in diameter and volume using the ellipsoid formula 4/3 π * ½ antero-posterior diameter * ½ latero-lateral diameter * ½ cranio-caudal diameter) and of functional changes (pattern flow) of parathyroid glands, and can therefore highlight a mismatch between morphological and biochemical data. Adding these data to biochemical profile, US/CD can evidence the inadequacy of pharmacological treatment or a poor compliance. It also allows the timing of surgery. According to international guidelines, a glandular volume ≥ 500 mm3 associated with i-PTH levels >700 pg/ml no longer controlled by pharmacological treatments are an indication for parathyroidectomy. The presence at US of a hyperplastic gland with a diameter > 1 cm is considered by several authors [99,125] a parameter of autonomous growth and irreversibility of sHPT and therefore it is an indicator of parathyroidectomy. Color-Doppler for the study of the parathyroid glands is performed with multifrequency linear probes (7.5 to 12 MHz), while in obese patients with goitre it can be used a low frequency probe (5 to 7.5 MHz). In clinical practice, we recommend an US examination when serum i-PTH >500 pg/ml.
Scintigraphy is essential for the presurgical localization of ectopic
glands. The role of scintigraphy in the preoperative localization of hyperplastic parathyroid glands has been debated for a long time. In the past decades, the diagnostic accuracy of imaging techniques was often inadequate, so Doppman wrote the aphorism that "the only preoperative localization study Indicated in a patient with untreated HPT is locating an experienced parathyroid surgeon" [126]. Technological advances in recent years, however, led to a significant improvement in diagnostic accuracy of these methods, but with higher costs, so the discussion about their use in clinical practice is still open. The first tracer used since the 80s was the 201
parathyroid tissue so it is necessary to use some technical tricks to detect it: the double-tracer technique with subsequent subtraction of the image needs the administration 201Thallium and another thyroid tissue specific radiotracer (99mTecnetium). This technique, however, had a very poor accuracy (around 50%), so it has a limited use. The recent introduction of more selective tracers in clinical practice, as the 99m-Tc-Sestamibi and 99m Tc-tetraphosmine scintigraphy and SPECT-CT technology have led to an improvement in sensitivity and specificity. 99mTc-Sestamibi (6-methoxy-butyl-isonitrile) is a cationic complex which shows a strong tropism for cells and glands with a high metabolic turnover. It is accumulated in the thyroid and parathyroid tissue in few minutes [127], but it is released more rapidly from the thyroid than from parathyroid glands: a comparison between the recorded image at 15 minutes and at 3 hours after administration allows evidencing the hyperfunctioning parathyroid tissue. SPECT provides information about the three-dimensional location of the lesion and about the topographical relationships with surrounding anatomical structures, so it is very useful (sensitivity greater than 90%) in difficult ectopic locations [128].
Computed Tomography (CT) is used in the localization of ectopic
mediastinal parathyroid glands. Adenomas and hyperplasia are usually hypodense compared to thyroid tissue; adenomas generally have a good contrast enhancement that makes them isodense. Nevertheless, the differential diagnosis between thyroid nodules, parathyroid glands and cervical lymph nodes can be difficult.
Magnetic Resonance (MR) has the same indications of CT.
Hyperplasia and adenomas are isointense with the thyroid and muscles on T1-weighted sequences, a hypointense on T2-weighted sequences and in those with fat suppression.
BACKGROUND OF THE STUDY
Diagnosis and treatment of sHPT are based on the dosage of serum i-PTH, Calcium (Ca), phosphorus (P) and vitamin D. Unfortunately, the optimal values of PTH in maintenance hemodialysis (MHD) patients and peritoneal dialysis (PD) are unknown [89]. According to the K-DOQI guidelines, the optimal values are 150-300 pg/ml (n.v. 10-70 pg/ml) so the prevalence of sHPT ranges from 20-30% [18] to 47-50% [19,20]. Recent K-DIGO guidelines indicate that in MHD i-PTH values from two to nine times the upper limit of normal i-PTH values
However, the dosage of i-PTH can cause diagnostic problems due to the heterogeneity of the molecule, the circadian and pulsatile secretion, the preanalytical variability (sampling times and storage conditions of blood sample) and the analytical variability (calibration differences between the different assay methods) [129]. Finally, the use of phosphate binders, vitamin D/analogues and calcimimetics reduces the secretion of PTH changing the serum levels of i-PTH regardless of the degree of hyperplasia.
are acceptable (140-630 pg/ml) [89].
So, in clinical practice biochemical data do not accurately represent the degree of sHPT, the glandular involvement and the structural changes that hyperplastic glands can show due to a response to therapy [116]. In this context, the role of US/CD is to integrate the biochemical diagnosis by defining the number, location, volume and structure of hyperplastic glands. Scintigraphy should be performed when US/CD is constantly negative, but the persistent increase of i-PTH is suspicious of an ectopic gland [130,131].
AIM OF THE STUDY
The primary objective of this prospective observational study was to trace the epidemiology of sHPT in a population of MHD patients, based on biochemical parameters of mineral metabolism (serum i-PTH, Ca, P, CaXP product, Alkaline phosphatase)
Secondary objectives of the study were: 1) to integrate the biochemical and pharmacological data with the morphological data obtained with US/CD, 2) to study the relationship between biochemical sHPT and glandular volume, 3) to establish the correlation between serum i-PTH and total gland volume.
and on morphological and perfusional characteristics, in order to characterize the glandular hyperplasia with US/CD and to study the relationship between biochemical parameters and gland volume.
PATIENTS AND METHODS
Patients
We enrolled 395 patients, 269 M and 126 F, 380 in MHD and 15 in PD, enrolled in the Nephrology and Dialysis Units of the Hospitals of Pisa, Pontedera, Lucca, Barga, Castelnuovo Garfagnana, Versilia and Grosseto). All patients underwent US/CD examination of the neck. The examination was performed by two single-blinded experienced operators.
Study design
The evaluation of US/CD parameters was preceded by the collection in a case report format of data on age, sex, renal disease, comorbidities (cardiovascular, cerebrovascular, peripheral vascular diseases, cancer, infections), the presence of metabolic disease (diabetes mellitus, dyslipidemia) and risk factors such as smoke. Were recorded the parameters of mineral metabolism: serum calcium, phosphorus, Ca x P product, alkaline phosphatase and i-PTH and those of lipid metabolism (total cholesterol, HDL, LDL, triglycerides). In each patient we evaluated the uremic age (pre dialytic and dialytic age) and other parameters: Kt/V values (dialytic efficacy), systolic, diastolic and differential blood pressure measured before the dialysis session, weight after the dialysis session, calcium-phosphorus metabolism drug therapy.
Methods
US/CD of the neck for the assessment of the parathyroid glands was performed with a portable device (My-Lab25, Esaote Biomedica, Genoa) and digital equipment (Logiq 9TM, General Electric Medical Systems, USA), with a variable frequency linear probe 9-14 MHz (axial and lateral resolution of 0.2 mm). Glandular flows were sampled with a transmission frequency of 3.8 MHz, a pulse repetition frequency (PRF) <800 kHz, a
steering of 20-30° and a medium-high colour gain (between 45 and 50 in a scale of 0-60).
Patients were examined with the neck hyperextended in a supine decubitus. The anatomical location of the parathyroid glands was determined by a sequence of transverse, longitudinal and oblique scans in the laterocervical region, from the corner of the jaw to the superior mediastinum. The mediocervical transverse scan was essential to identify the thyroid lobes and isthmus. The same transverse scan along the latero-cervical line was used to identify the usual landmarks (sternocleidomastoid muscle, prethyroid muscles, epiaortic vessels, oesophagus, prevertebral muscles). Longitudinal and oblique scans were used to define the anatomical relationships with the homolateral thyroid lobe, the oesophagus, the long muscles of the neck, the laterocervical vessels and the thyroid and cricoid cartilages. Additional oblique scans were carried out on the jugular fossa to identify the inferior parathyroid glands, sometimes located in the superior mediastinum or in the horn of the thymus. The inferior thyroid artery in the mid-cervical tract was also located to define vascularization of parathyroid glands.
B-Mode and color Doppler parameters
High-resolution US/CD is unable to distinguish normal parathyroid glands from the thyroid gland. This limit is related to the location, size, but also the quantity of stromal adipose tissue that makes the glands hyperechoic. Hyperplastic glands, however, are rich in oxyphil and main cells [6,9] so they become hypoechoic and more easily distinguishable. In our study we considered hyperplastic all glands with hypoechoic structure, outlined from the thyroid by a thin hyperechoic edge and with a long axis >5 mm. Gland volume was derived from the linear diameters using the irregular ellipsoid formula (4/3 π x ½ antero-posterior diameter x ½ latero-lateral diameter x ½ cranio-caudal diameter).
The correlation between total glandular weight measured after surgical excision and PTH serum levels is good (r2 = 0.081) but not high [132]. Our unpublished data [133] show a linear correlation between i-PTH and
calculated gland volume in patients not treated until the total gland volume does not exceed 2000 mm3. In any case, these data are doubtful because the measurement of volume or weight after glandular excision is not a common practice. Studies that correlate glandular volume with histological features of excised glands show that glands with a volume <500 mm3 are affected mainly by diffuse or multinodular polyclonal hyperplasia, while glands with a volume >500 mm3
At CD, hyperplastic glands showed a widespread increase in vascularization as confirmed by recent histological studies [134]. Three different vascularization patterns were identified: type 1) glands with no Doppler signal; type 2) hypovascularized glands with a poor or weak Doppler signal, represented by occasional colour spots in the hilar/endonodular region; type 3) hypervascularized glands, having an enlarged feeding artery at the hilum, a peripheral arc of vascularity and/or ray-like endonodular vessels. Where possible, we performed spectral analysis of the Doppler signal at the hilum and we determined peak systolic velocity (PSV) and resistance index (RI).
in 80% of cases are affected by monoclonal nodular hyperplasia [101].
There was no correlation between perfusion patterns and glandular function. A hyperfunctioning gland with a volume >500 mm3 can appear hypovascularized or avascularized, due to inadequate sampling (artefactual, deep location). In general, glands with diffuse hyperplasia and a volume <500 mm3 have more often a hypovascular pattern, while glands with nodular hyperplasia and a volume >500 mm3 were hypervascular in 78-80% of cases. Therefore, although vascular pattern is a semiquantitative parameter and it is not correlated with i-PTH, it is certainly an expression of increased volume and hyperfunction of the parathyroid gland.
Statistical analysis
For the statistical analysis of data, to evaluate the difference between the means of two populations, we used the one-tailed z test since the variances of the populations compared were known and samples were large
enough (n1 and n2 >30). The level of significance α of the test used to reject the null hypothesis (H0) was p ≤ 0.05. The statistical program used was Microsoft Excel ®-section Data Analysis. To assess the degree of homogeneity of the measurements made by two different operators we used the intra-class correlation coefficient (ICC). The intra-observer ICC for the measurement of gland volume was 0.95, while the inter-observer ICC was 0.91.
RESULTS
We enrolled 395 patients (269 m, 126 f) aged 68.2 ± 14.0 (m ± SD). Mean dialytic age was 62.9 ± 71.3 months and mean predialytic CKD age was 91.4 ± 82.7 months. Clinical and biochemical parameters of the population are shown in Table 1, while factors causing CKD are shown in
Figure 1.
Table 1. Clinical and biochemical parameters of the population studied.
N° of patients (m, f) 395 (269/126)
Mean age (years) (M ± SD) 68.2±14.0
Dialytic age (months) (M ± SD) 62.9 ± 71.3
Predialytic CKD age (months) (M ± SD) 91.4 ± 82.7
kt/V 1.4 ± 0.3 BMI (kg/m2) (M ± SD) 24.9 ± 6.0 Total cholesterol (mg/dl) (M ± SD) 166.2 ± 40.1 HDL (mg/dl) (M ± SD) 42.3 ± 13.3 LDL (mg/dl) (M ± SD) 98.8 ± 36,1 Triglycerides (mg/dl) (M ± SD) 165.7 ± 103.3 Ca (mg/dl) (n.v. 8.5-9.5) (M ± SD) 8.9 ± 0,8 P (mg/dl) ( n.v. 3.5-5.5) (M ± SD) 4.7 ± 1.5 Ca x P (mg2/dl2) ( n.v. <55) (M ± SD) 41.8± 15.0 ALP (IU/l) ( n.v. 130) (M ± SD) 133.5 ± 93.1 PTH (pg/ml) (n.v. 150-300) (M ± SD) 281.9 ± 233.4 SBP (mmHg) (m ± SD) 136±22 DBP (mmHg) (m ± SD) 70±12 Differential BP (mmHg) (m ± SD) 64±21
Fig. 1. Causes of CKD in the population studied.
108 patients (27.3%) had diabetes mellitus. Cardiovascular comorbidities are shown in Figure 2. 145 patients (36.7%) had one or more symptoms/signs of coronary artery disease [15 (10.3%) patients at least one episode of angina pectoris, 56 (38.6%) at least one episode of acute myocardial infarction (AMI), 31 (21.3%) signs and symptoms of chronic ischemic heart disease, and 43 (29.6%) patients had one or more cardiac manifestations of coronary artery disease ]. 110 patients (27.8%) had a history of cerebrovascular disease [11 (10%) at least one episode of TIA, 23 (21%) at least one episode of ischemic or hemorrhagic stroke, 42 (38%) US/CD signs of carotid stenosis or signs of previous tromboendoarteriectomy (TEA), 12 (10.9%) clinical and imaging (CT, MR) signs of chronic ischemic encephalopathy and 22 (20%) had more manifestations of cerebrovascular disease]. 85 (21.5%) patients showed signs of ischemic lower limb arteriopathy. The blood pressure values were in the normal range in 95 patients (23.9%). 300 (76.1%) patients were hypertensive in mono/polytherapy.
Fig. 2. Cardiac and cerebrovascular comorbidities of the population studied.
Biochemical parameters of mineral metabolism are shown in Table 1. Serum Ca was 8.9 ± 0.8 mg / dl, P 4.7 ± 1.5 mg / dl, Ca x P product 41.8 ± 15.0 mg2/dl2, ALP 133.5 ± 93.1 IU/l and i-PTH 281.9 ± 233.4 pg/ml. According to the reference values of the K-DOQI guidelines, 252 patients showed values of i-PTH ≤ 300 pg/ml (144.1 ± 78.9 pg/ml) and 143 patients values of i-PTH> 300 pg/ml (526.6 ± 215.9 pg/ml). Based on K-DOQI international guidelines, 143 patients had sHPT (prevalence 36.2%). Table
2 summarizes the parameters of mineral metabolism and the statistical
differences between the two groups. Serum values of Ca did not differ significantly while P, Ca x P product and ALP were significantly different (P 4.6 ± 1.4 mg/dl vs 4.9 ± 1.7 mg/dl, p = 0.03, Ca x P 40.5 ± 13.6 mg2/dl2 vs 44.4 ± 16.6 p = 0.01, ALP 124.7 ± 87.2 IU/l vs 148.9 ± 101.1 IU/l, p=0.01)
i-PTH ≤ 300 pg/ml i-PTH > 300 pg/ml p N° of patients (M/F) 252 (175/77) 143 (94/49) Age (years) (m ± SD) 69.2±14.1 66.7 ± 14.2 NS Dialytic age (months) (m ± SD) 60.3 ± 76.2 67.5 ± 61.9 NS Predialytic CKD age (months) (m ± SD) 82.7 ± 77.1 105.7 ± 89.5 0.01 s-Ca (mg/dl) (m ± SD) 9.0 ± 0.8 9.0 ± 1.0 N.S. s-P (mg/dl) (m ± SD) 4.6 ± 1.4 4.9 ± 1.7 0.03 Ca x P (mg2/dl2) (m ± SD) 40.5 ±13.8 44.4 ± 16.6 0.01 ALP (IU/l) (m ± SD) 124.7 ± 87.2 148.4 ± 101.1 0.01 iPTH (pg/ml) (m ± SD) 144.1 ± 78.9 526.6 ± 215.9 <0.001
Table 2: Clinical and biochemical characteristics of the two groups of
patients (i-PTH<300 pg/ml and i-PTH>300 pg/ml).
Each dialysis unit used its own drug treatment schema for the control of mineral metabolism, so drug associations were difficult to compare. 110 patients (28%) receive phosphate binders alone, 119 (30%) received phosphate binders with vitamin D analogues, 36 (9%) vitamin D analogues alone, 8 (2%) calcimimetic monotherapy, 12 (3%) calcimimetics and phosphate binders, 23 (6%) calcimimetic and vitamin D/analogues, 36 (9%) conventional therapy and cinacalcet.
Finally, 51 (13%) patients did not receive any therapy for calcium-phosphorus metabolism (Fig. 3).
Fig. 3. Drug treatment in the population studied.
US/CD showed 173 parathyroid glands in 108 patients (prevalence of moprhological sHPT: 27.3%): 70 (64.9%) patients showed i-PTH >300 pg/ml and 38 (35.1%) i-PTH ≤300 pg/ml. 73/287 (25.4%) patients had iPTH values >300 pg/ml but they didn’t show hyperplastic parathyroid glands at US/CD.
214/287 (74.6%) showed values of i-PTH ≤ 300 pg/ml without hyperplastic parathyroid glands at US/CD.
63 patients (58.3%) had only one hyperplastic gland, 28 (26%) two hyperplastic glands, 14 (13%) three hyperplastic glands, 3 (2.7%) four hyperplastic glands. Eleven parathyroid glands (6%) had an ectopic location (antero-superior mediastinum and jugular fossa).
The structure of hyperplastic glands was almost diffusely hypoechoic but in some cases it showed hyperechoic septa within the hypoechoic parenchyma. Only in patients treated with calcimimetics and conventional therapy we found cystic-like lesions of different diameters associated with a widespread hypovascular pattern.
Calculated gland volume ranged from 12.5 to 3979 mm3 and the mean glandular volume was 257.6 ± 419.6 mm3
N° of parathyroid glands
(m ± SD). The US/CD characteristics of the glands are shown in Table 3.
173
Mean glandular volume (mm3) 257.6 ± 419.6
N° glands with Volume ≤500 mm3 153
N° glands with Volume >500 mm3 20
Vascular pattern of glands with Volume ≤ 500 mm3
Pattern 0 42/153 (27.5%) Pattern 1 55/153 (36%) Pattern 2 54/153 (35.5%)
Vascular pattern of glands with Volume > 500 mm3
Pattern 0 1/20 (5 %) Pattern 1 5/20 (25 %) Pattern 2 14/20 (70 %)
Table 3. US/CD characteristics of parathyroid glands.
18 patients had at least a gland with a volume >500 mm3 and in 90 patients each gland had volumes ≤500 mm 3. The vascular pattern of hyperplastic glands varied depending on gland volume: 24.9% of glands showed a complete absence of vascularization, 34.7% of glands showed a hypovascular pattern, while 40.4% of glands had a hypervascular pattern. Glands with a small volume showed no vascularisation or rare vascular signals (pattern 0 or 1), while 14 of the 20 glands with a volume >500 mm3
Based on the presence/absence of hyperplastic parathyroid glands, the study population could be divided into two groups: Group 1 (patients without morphological sHPT) and Group 2 (patients with morphological sHPT). The two groups differed significantly in age (p <0.001), dialytic age (p = 0.0002) and predialytic CKD age (p = 0.0003). Serum calcium and showed a predominantly hypervascular pattern (pattern 2) with dilation of the arteries, 5 glands showed a hypovascular pattern and only 1 did not show vascularization. These data suggest a possible correlation between nodular hyperplasia evidenced by US/CD and hypervascularity. In other words, vascular pattern could be an indirect parameter of the severity of glandular hyperfunction.
ALP showed no significant differences between the two groups, while serum P, Ca x P product and i-PTH were significantly different (P 4.5 ± 1.5 vs 5.1 ± 1.6 mg/dl, p = 0.001; Ca x P product 40.1 ± 13.5 vs 46.6 ± 17.7 mg2/dl2, p =0.0001 and i-PTH 226.7 ± 184.4 vs 430.2 ± 282.8pg/ml, p < 0.0001). Table 4 summarizes the differences between the two subgroups and Fig. 4 and Fig. 5 show the differences in etiology of CKD between the two groups. Finally, patients with parathyroid hyperplasia and i-PTH ≤300 pg/ml showed a mean glandular volume of 169.8 ± 251.4 mm3, whereas patients with parathyroid hyperplasia and i-PTH >300 pg/ml had a mean glandular volume of 295.4 ± 468.7 mm3 (p = 0.07).
Table 4. Clinical and biochemical characteristics of patients with and
without morphological sHPT. Absence of morphological sHPT Morphological sHPT p N° of patients (m/f) 287 (201/86) 108 (68/40)
Mean age (years) (m±SD) 70.4 ± 13.4 62.6± 14.2 <0.001
Mean dialytic age
(months) (m±SD) 54.9 ± 69.5 84.4 ± 72.0 0.0002
Mean predialytic CKD age
(months) (m±SD) 82.3 ± 75.6 115.3 ± 95.0 0.0003 s-Ca (mg/dl) (m ± SD) 8.9 ± 0.8 9.1 ± 0.9 N.S. s-P (mg/dl) (m ± SD) 4.5 ± 1.5 5.1 ± 1.6 0.001 Ca x P (mg2/dl2) (m ± SD) 40.1 ± 13.5 46.6 ± 17.7 0.0001 ALP (IU/l) (m ± SD) 129.6 ± 90.2 143.4 ± 99.8 N.S. iPTH (pg/ml) (m ± DS) 226.7 ± 184.4 430.2 ± 282.8 < 0.0001 PAS (mmHg) (m ± DS) 135.8 ± 20.7 135.0 ± 23.9 N.S. PAD (mmHg) (m ± DS) 69.2 ± 11.3 70.8 ± 14.1 N.S.
Fig. 4. Causes of CKD in patients without morphological sHPT.
Fig 5. Causes of CKD in patients with morphological sHPT.
The correlation between i-PTH serum levels and total glandular (Fig.
6) showed a linear trend regardless of the current treatment. The equation of
the linear regression line was glandular volume = 0.8995 * (i-PTH) + 15.3. The Pearson correlation coefficient was R = 0.3694, and R2 = 0.1365 (p = 0.001).
